The patent and scientific literature includes various viral vector systems, uses therefor, and exogenous DNA for expression of protein by such systems, as well as uses for such proteins and uses for products from such proteins.
For instance, recombinant poxvirus (e.g., vaccinia, avipox virus) and exogenous DNA for expression in viral vector systems can be found in U.S. Pat. Nos. 5,174,993 and 5,505,941 (e.g., recombinant avipox virus, vaccinia virus; rabies glycoprotein (G), gene, turkey influenza hemagglutinin gene, gp51,30 envelope gene of bovine leukemia virus, Newcastle Disease Virus (NDV) antigen, FelV envelope gene, RAV-1 env gene, NP (nudeoprotein gene of Chicken/Pennsylvania/1/83 influenza virus), matrix and preplomer gene of infectious bronchitis virus; HSV gD; entomopox promoter, inter alia), U.S. Pat. No. 5,338,683, e.g., recombinant vaccinia virus, avipox virus; DNA encoding Herpesvirus glycoproteins, inter alia; U.S. Pat. No. 5,494,807 (e.g., recombinant vaccinia, avipox; exogenous DNA encoding antigens from rabies, Hepatitis B, JEV, YF, Dengue, measles, pseudorabies, Epstein-Barr, HSV, HIV, SIV, EHV, BHV, HCMV, canine parvovirus, equine influenza, FeLV, FHV, Hantaan, C. tetani, avian influenza, mumps, NDV, inter alia); U.S. Pat. No. 5,503,834 (e.g., recombinant vaccinia, avipox, Morbillivirus [e.g., measles F, hemagglutinin, inter alia]); U.S. Pat. No. 4,722,848 (e.g., recombinant vaccinia virus; HSV tk, glycoproteins [e.g., gB, gD], influenza HA, Hepatitis B [e.g., HBsAg], inter alia); U.K. Patent GB 2 269 820 B and U.S. Patent No. 5,514,375 (recombinant poxvirus; flavivirus structural proteins); WO 92/22641 (e.g., recombinant poxvirus; immunodeficiency virus, inter alia); WO 93/03145 (e.g., recombinant poxvirus; IBDV, inter alia); WO 94/16716 and U.S. application Ser. No. 08/184,009, filed Jan. 19, 1994 (e.g., recombinant poxvirus; cytokine and/or tumor associated antigens, inter alia); and PCT/US94/06652 (Plasmodium antigens such as from each stage of the Plasmodium life cycle).
Baculovirus expression systems, exogenous DNA for expression therein, and purification of recombinant proteins therefrom can be found in Richardson, C. D. (Editor), Methods in Molecular Biology 39, "Baculovirus Expression Protocols" (1995 Humana Press Inc.) (see, e.g., Ch.18 for influenza HA expression, Ch.19 for recombinant protein purification techniques), Smith et al., "Production of Huma Beta Interferon in Insect Cells Infected with a Baculovirus Expression Vector," Molecular and Cellular Biology, Dec., 1983, Vol. 3, No. 12, p. 2156-2165; Pennock et al., "Strong and Regulated Expression of Escherichia coli B-Galactosidase in Infect Cells with a Baculovirus vector," Molecular and Cellular Biology Mar. 1984, Vol. 4, No. 3, p. 399-406; EPA 0 370 573 (Skin test and test kit for AIDS, discussing baculovirus expression systems containing portion of HIV-1 env gene, and citing U.S. application Ser. No. 920,197, filed Oct. 16, 1986 and EP Patent publication No. 265785).
U.S. Pat. No. 4,769,331 relates to herpesvirus as a vector.
There are also poliovirus and adenovirus vector systems (see. e.g., Kitson et al., J. Virol. 65, 3068-3075, 1991; Grunhaus et al., 1992, "Adenovirus as cloning vectors," Seminars in Virology (Vol. 3) p. 237-52, 1993; Ballay et al. EMBO Journal, vol. 4, p. 3861-65; Graham, Tibtech 8, 85-87, April, 1990; Prevec et al., J. Gen Virol. 70, 429-434).
PCT WO91/11525 relates to CAV2 modified to contain a promoter-gene sequence within the region from the SmaI site close to the end of the inverted terminal repeat region up to the promoter for the early region 4 (E4).
CAV, and particularly CAV2, has numerous problems. Several of these problems are discussed below. A significant problem is that the CAV genome can only accept a limited amount of exogenous DNA. That is, only a limited amount of exogenous DNA can be inserted into the CAV genome. Thus, CAV is "insert size limited" and therefore presents a significant problem which must be addressed if CAV is to be a useful vector for a constellation of cloning and expression applications.
The efficient transmission of many viral infections via the oronasal route has provided the impetus for assessing the efficacy of viral vector-based vaccine candidates via the same route. However, since the spread of most live replicating vaccines within the vaccinee and their spread to or contacts with the general environment are well documented (for examples see Schwartz et al., 1974, Mueller et al., 1989, Oualikene et al., 1994), the choice of an adequate viral vector is not obvious.
To address legitimate safety concerns, vector selection preferably involves consideration of characterized live attenuated vaccines as the apparent safety thereof is established. For vaccination of humans, various vectors based on replicating live attenuated viruses are under consideration. To date, there are documented approaches based on human adenoviruses (HAVs) serotype 4 and 7 (Lubeck et al., 1989, Chanda et al., 1990, Chengalvala et al., 1991, 1994, Hsu et al., 1994 ), influenza viruses (for a review Garcia-Sastre and Palese, 1995) and poliovirus and related viruses (for a review Girard et al., 1995).
In the field of veterinary medicine, several vectors based on replicating live attenuated viruses are currently being analyzed with the objective to apply those recombinant vectors as vaccines either parenterally or via the natural route of infection, thereby stimulating local protection. Among the best characterized at this point are members of the poxviridae family [e.g., fowlpox-based vectors (Edbauer et al. 1990, Taylor et al., 1995 and ref. therein)], herpesviridae family [e.g., pseudorabies virus-based vectors (Sedegah et al. 1992, Mettenleiter et al. 1994, Hooft van Iddekinge et al., 1996 and ref. therein), turkey herpes virus-based vectors (Ross et al. 1993, Darteil et al. 1995 and ref. therein), feline herpes virus-based vectors (Cole et al., 1990, Wardley et al., 1992, Willense et al., 1996), infectious laryngotracheitis virus-based vectors (Guo et al., 1994), bovine herpes virus-based vectors (Kit et al. 1991)] and to a lesser extent members of the Adenoviridae family [bovine adenovirus 3-based vectors (Mittal et al., 1995)].
The canine species provides an appropriate model for oronasal immunizations. As such, the canine adenovirus serotype 2 (CAV2) for which attenuated vaccinal strains exist that can be safely administrated either parenterally or via oronasal route, provides a viable immunization vehicle for canine vaccination. Canine distemper virus (CDV) infection of dogs provides a good example of a respiratory infection in this target species. Further, a relatively direct experimental CDV challenge system is accessible and allows a direct comparison between CAV2 based-vaccine candidates and previously developed classical CDV vaccines.
CAV2 was first isolated from an outbreak of upper respiratory tract infection in dogs by Ditchfield et al. (1962). Since then, the virus has been isolated from the respiratory tract of dogs with respiratory diseases both in the US and in Europe (Binn et al. 1967, Appel and Percy, 1970, Assaf et al. 1978, Danskin 1973). Experimental studies have resulted in mild respiratory disease following aerosol inoculation of CAV2 (Swango et al. 1970, Appel , 1970). Several CAV2-based vaccines have been developed and extensively used worldwide for the vaccination of puppies and adult dogs. Immunization with CAV2 has even been shown to protect against an experimental challenge exposure with a serologically related strain of CAV1, which is fatal to non-vaccinated dogs (Fairchild and Cohen, 1969, Appel et al. 1973, Bass et al. 1980). The apparent safety of CAV2 as a vaccine has been well evidenced by the lack of vaccine-induced and vaccine-associated complications in dogs and other animal species including man during its 30 years of utility. Further, results from field serological surveys indicate that many wild animals (foxes, raccoons, skunks and mongooses) are asymptomatically exposed to CAV2 or to an antigenically related virus infection (Summer et al., 1988). A vaccinal strain of canine adenovirus serotype 2 (CAV2), therefore, provides a unique example of a safe replication-competent, host-restricted virus which can be considered for the derivation of effective vector-based vaccine candidate for vaccination, especially of dogs.
HAVs have been shown to be valuable mammalian cell expression vectors (for a review see Graham et al. 1988) and are currently being evaluated both as recombinant viral vaccine candidates (for reviews see Randrianarison-Jewtoukoff and Perricaudet 1995, Imler 1995) and as vectors for gene therapy (for reviews see Perricaudet and Perricaudet 1995). There are two major groups of HAVs, and a third, less explored, group of recombinant HAVs.
The first group of these adenovirus vectors corresponds to replication-incompetent recombinant adenoviruses which are based on viruses deleted of their E1 region. The E1 region encodes proteins which are essential for virus replication in tissue culture. It has, however, been demonstrated that replication-incompetent recombinant adenoviruses deleted of their E1 region can be propagated in the 293 cell line (Graham et al., 1977) which constitutively expresses the E1 region (Haj-Ahmad et al., 1986).
Deletion of the E1 region not only increases the amount of foreign DNA which can be inserted into HAVs, but also limits their replication in human cells and thus considerably improves the safety characteristics of the corresponding recombinant HAVs in humans. Most of the HAV-based vaccine candidates against veterinary and humans pathogens are currently based on E1-deleted vectors. Despite their limited replicative capacity, protection data in challenge experiments have been described (Prevec et al., 1989, McDermott et al., 1989, Lubeck et al., 1989, Eloit et al., 1990, Ragot et al., 1993, Wesseling et al., 1993, Both et al., 1993, Gallichan et al., 1993, Hsu et al., 1994, Breker-Klasser et al., 1995). The property of inducing a protective immune response even in the absence of vector replication is shared by other host restricted viral vectors, the most promising of which being the canarypox virus-based vector ALVAC (Taylor et al., 1991, see Perkus et al., 1995 for a review).
When the goal is a replication competent adenovirus vector, the use of the E1 region as an insertion site is thus not desirable; and, the E1 region therefore has heretofore had deficiencies and presented problems. These deficiencies and problems are compounded when a replication competent adenovirus displaying safety characteristics with respect to humans is desired. In particular, while the E1 region deletion in HAVs may limit replication in human cells and improve safety characteristics with respect to humans, as discussed below, the possibility of recombination between E1 transformed cell lines and E1 deleted recombinant adenoviruses has been documented and thus the safety profile of E1 transformed cell lines appears questionable, thereby rendering any benefit from using E1 region deleted adenoviruses potentially illusory and exascerbating deficiencies and problems heretofore in the use of E1 region deleted adenoviruses (since propagation of E1 region deleted adenoviruses is in cells which constitutively express the El region).
The second group of adenovirus vectors corresponds to recombinant adenoviruses which are replication-competent in human cells but replication-incompetent in most non-human animal cells. Those viruses are characterized by a substitution of part of the E3 region with foreign gene expression cassettes. The E3 region has been shown to be non-essential both in vitro and in vivo for infectious virus formation (Kelly and Lewis 1973, Kapoor et al., 1981, Morin et al., 1987, Lubeck et al., 1989). Numerous recombinant HAVs have therefore been generated by replacement of part of the E3 region (Morin et al., 1987, Chengalvala et al., 1991, 1994, Prevec et al., 1989, Johnson et al., 1988, Lubeck et al., 1989, Dewar et al., 1989, Natuk et al., 1993, Hsu et al., 1994).
However, since proteins encoded by the E3 region have been shown to alter various aspects of the host immune responses (for a review see Wold and Gooding 1991), E3 deletion may have some impact on the pathogenic profile of corresponding recombinant viruses. Indeed, it has been demonstrated in a cotton rat model that deletion of the E3 region from HAV serotype 5 increases virus pulmonary pathogenicity (Ginsberg et al., 1989). However, it has also been demonstrated that a recombinant bovine Ad3, partially deleted within its E3 region, produces lesions in cotton rats similar to those observed with the parental wt bovine Ad3, therefore suggesting that safety of bovine Ad3-based vectors may be sufficient for the derivation of live recombinant virus vaccines for cattle (Mittal et al., 1996).
These results also show that the impact of deletions within the E3 region of any specific adenovirus should be considered on a case-by-case approach.
The CAV2 E3 region has been identified and characterized previously (Linne, 1992). However, based on the available published data (Linne 1992), the precise definition of an insertion site in the CAV2 E3 region is not obvious. DNA sequence analysis revealed that the organization of the CAV2 E3 region differs significantly from that described for HAVs. The human adenovirus E3 region corresponds to a stretch of at least 3 kbp containing at least 8 open reading frames (orf) whereas the CAV2 E3 region is only 1.5 kbp long and contains only 3 orfs. None of these orfs have a significant level of homology with HAV E3 orfs. From such preliminary comparative analyses, it appears reasonable to speculate that human and canine adenoviruses genomes have evolved differently.
The definition of an insertion site within the CAV2 E3 region is further complicated by the complex splicing and polyadenylation pattern which characterizes the adenovirus family (for a review Imperiale et al., 1995). RNA splicing donor and aceptor sites localized within the E3 region may be important for the maturation of several essential mRNAs even though their coding sequences are localized outside of the E3 region.
Further, since the E3 region is located within a genome region of high transcriptional activity (for a review Sharp et al., 1984), the insertion of foreign DNA at this site has a potential detrimental impact on the biology of the recombinant virus. Additionally, the E3 region is located downstream of the major late promoter (MLP), where interference between transcription of recombinant gene and transcription initiated at the MLP has been demonstrated (Zu et al., 1995).
Problems in the art to be addressed therefore include: minimizing phenotypic alterations of the recombinant virus, and the definition of an insertion site in a less transcriptionnally active region. And, in general, it can be said that the E3 region presents problems in the art which should be addressed.
The less explored third group of recombinant HAVs is based on the insertion of recombinant DNA between the right inverted terminal repeat (ITR) and the E4 promoter. The ITRs contain sequences which are essential for viral DNA replication and efficient packaging of the viral genomic DNA. While a region between the right inverted terminal repeat (ITR) and the E4 promoter may accommodate exogenous DNA sequences (Saito et al., 1985, Chanda et al., 1990), adenoviruses-based vectors have severe limitations in the amount of foreign DNA they can carry, as the packaging capacity of recombinant hAd5 is limited to a genome of approximatively 105% of the wild-type genome (Bett et al. 1993); thus presenting a problem in the art.
While the region between the right ITR and the E4 region may represent an additional insertion site candidate for the generation of CAV2 recombinant viruses, and PCT WO 91/11525 may relate to a SmaI site close to the leftward extremity of the ITR as a potential insertion site. Contrary to the teachings of WO91/11525, there appears to be an upper limit for insertion at this site as Applicant attempted insertions at this site and was able to insert a 400 bp DNA fragment, but larger insertions such as a fragment approximately 1 kbp repeatedly failed to be introduced into the site. Hence, a problem in the art is the utility of this site.
Therefore, the E4 promoter region has heretofore had deficiencies and presented problems.
Initial characterization of the CAV2 genome at the molecular level has been described in the literature. Restriction analysis of several strains of both CAV2 and CAV1 (Jouvenne et al., 1987, Macartney et al., 1988, Spibey and Cavanagh 1989) and sequence analysis of the corresponding E1, E3 and ITRs regions have been reported (Cavanagh et al., 1991, Linne 1992). Although the overall genomic organization of canine adenoviruses is similar to those described for other Adenoviridae family members, the precise organisation of CAV2 genomic E3 region is unique.
Accordingly, one cannot merely extrapolate from one member to another member of the Adenoviridae family, thereby providing yet another problem in the art.
Further still, when addressing any or all of the aforementioned deficiencies or problems, it would be preferred to avoid any dependence on an endogenous promoter like the E3 or the MLP promoters. However, the pattern of expression of the recombinant gene may be a critical parameter in the overall expression and ergo in the efficacy of the recombinant in a vaccine or immunological composition (Darteil et al., 1995, Xu et al., 1995, Hooft van Iddekinge et al., 1996).
Several cellular and viral promoters have been involved in the derivation of recombinant HAVs. Among the best characterized are b-actin, SV40 early, SV40 late, hAD MLP, and hCMV-IE (Zu et al., 1995). The hCMV-IE promoter may have promise as an upstream regulatory region, since it is associated with the highest level and the longest persistence of recombinant protein expression in tissue culture. This promoter also appears to operate in almost every cell line tested thus far. A potential for cell type independent promoter activity can be regarded as a clear advantage.
It has been demonstrated that the hCMV-IE promoter can be transactivated by HAV infection (Gorman et al., 1989). The large size of this promoter (approximately 850 bp) is a problem with respect to the size limitations of recombinant CAV vector. Thus, one cannot merely extrapolate from past successes with this promoter to a recombinant CAV vector.
Adenoviruses are known to strongly repress the synthesis of cellular proteins after the onset of viral DNA replication (for a review Zhang and Schneider, 1993). Thus, replication-competent recombinant adenoviruses have heretofore had a potential for a strong limitation of the recombinant protein expression after the onset of DNA replication.
Similarly, Saito et al. (1985) demonstrated that a recombinant human adenovirus serotype 5 can produce high amounts of recombinant mRNA but that almost no recombinant protein is obtained.
Late adenovirus mRNAs are characterized by the presence of a tripartite leader (TPL) sequence in their 5' untranslated region (5'UTR). The presence of the TPL can be an important component of the translatability of late adenovirus mRNAs. Further, it has been demonstrated that in an hAd5 background, the presence of the TPL is a feature of the translational control of a recombinant SV40 T antigen expressed from adenovirus late promoter (Thummel et al. 1983).
Another important problem to address in the design of an expression cassette is the size of the polyadenylation signal.
Even still further, the problems in the art include establishing conditions to transfect CAV2 DNA into monolayers. The infectivity of purified naked adenovirus DNA is low. Using a calcium phosphate-based procedure, Graham and Van der Berg (1973) report a yield of 1 pfu/mg of purified DNA. This is not an efficient process for isolating recombinant viruses. Several approaches have been proposed to attempt to address this problem; but, none heretofore have fully addressed the problem, and particularly without raising additional issues such as safety.
For instance, DNA protein complexes have been purified and are reported to have an increased infectivity (5.times.10.sup.3 pfu/mg) (Sharp et al., 1976) over naked DNA. Similarly, covalently closed circles of adenovirus DNA have also been shown to be infectious (Graham, 1984).
A widely used procedure to derive recombinant HAVs is based on the utilization of the 293 cell line which has been transformed with the HAV E1 region (Graham et al., 1977). Previously, it has been reported that the derivation of bovine and canine adenovirus recombinants was dependent on the utilization of cell lines transformed with the corresponding adenovirus E1 region (PCT WO 91/11525, Mittal et al. , 1995a). However, since the genes encoded by the E1 region of some adenoviruses have been shown to contribute to the transformation of rodent cells (reviewed by Grand, 1987), the safety profile of E1 transformed cell line appear questionable. The presence of potent transactivators within the adenovirus E1 region (for a review Nevins, 1993) is also well established and further extends safety concerns which can be raised regarding E1 transformed cell lines.
Thus, transfection conditions independent of use of an E1 transformed cell line, especially with good yields, would be a significant advance in the art.
Accordingly, it is believed that a recombinant CAV, preferably a recombinant CAV2, having exogenous DNA inserted therein and a non-essential region or portion thereof deleted therefrom, especially such a CAV which is packaged as an infectious CAV with respect to cells in which CAV naturally replicates, or a CAV containing exogenous DNA within the E3 and/or the right end of the genome between the right ITR and the E4 transcription unit, and methods for making such recombinants, and uses for such recombinants, as described herein (above and below), has not been taught or suggested. Further, it is believed that a truncated transcriptionally active promoter for a recombinant virus or plasmid which comprises a region transactivated with a transactivating protein provided by the virus or a system into which the plasmid is inserted and the minimal promoter region of the promoter, an expression cassette comprising the promoter, and viruses or plasmids containing the promoter or expression cassette, have not been heretofore described or suggested. And, such a recombinant CAV and methods of making and using such a recombinant CAV, and such a promoter, expression cassette and viruses and plasmids containing the promoter or expression cassette present an advancement over prior recombinants, especially since as to humans CAV is a non-replicating vector and the promoter and expression cassette address insert size limits of recombinant viruses.